Solid-state conformation of anti-human immunodeficiency virus type-1

Contribution from the Medical Foundation of Buffalo, Inc., Buffalo, New York 14203, College of ... Administration Medical Center, Decatur, Georgia 300...
0 downloads 0 Views 1MB Size
2277

J . Am. Chem. SOC.1988, 110, 2277-2282

Solid-state Conformation of Anti-Human Immunodeficiency Virus Type- 1 Agents: Crystal Structures of Three 3’-Azido- 3’-deoxythymidine Analogues Patrick Van Roey,*+ Jeffrey M. Salerno,+ William L. Duax,+ Chung K. Chu,; Moon K. Alan,$ and Raymond F. Schinazii Contribution from the Medical Foundation of Buffalo, Inc., Buffalo, New York 14203, College of Pharmacy, University of Georgia, Athens, Georgia 30602, and Emory University and Veterans Administration Medical Center, Decatur, Georgia 30033. Received September 21, 1987

Abstract: The crystal structures of three anti-HIV 3’-azido-3’-deoxynucleosideshave been determined to gain conformational information for structureactivity studies. The compounds 3’-azido-3’-deoxythymidine (AZT), 3’-azido-2’,3’-dideoxyuridine (CS-87), and 3’-azido-2’,3’-dideoxy-5-ethyluridine (CS-85) are all active inhibitors of HIV- 1 replication. X-ray diffraction data for all three compounds were measured at 165 K. AZT: P21, a = 17.538 ( I ) A, b = 12.021 (1) A, c = 5.6435 (3) A. p = 95.596 (8)O, Z = 4, Rail= 0.038. CS-87: pZI, a = 9.7314 (8) A, b = 6.7517 (6) A, c = 8.3464 (7) A, p = 91.854 (7)O, Z = 2, Rall= 0.046. CS-85: a = 5.567 (5) A, b = 22.093 (9) A, c = 21.149 (4) A, 8, = 96.005 (9)O, Z = 8, Rail = 0.036. Large differences are observed in the conformations of the seven independent observations, including four different conformations for the glycosyl link and three furanose ring geometries. The azido group is nonlinear and has a preferred conformation trans to the C2’-C3’ bond (four molecules). The azido group of one CS-85 molecule is disordered. Two positions, cis to C2’-C3’ and 26’ from this position, are fully resolved. The three structures show extensive but distinct hydrogen-bonding patterns that include two different base-pairing geometries. AZT has one dimer formed by two molecules bonded by N3-02 hydrogen bonds. CS-85 has two dimers, one based on N3-02 hydrogen bonds and one based on N3-04 hydrogen bonds. The structure of CS-87 does not show base pairing but rather linear strands of molecules hydrogen bonded through the bases. The 5’-hydroxyl groups form additional strong hydrogen bonds with either the remaining carbonyl oxygens or 4’ ether oxygens. Strong correlations between intramolecular conformational differences and intermolecular interactions are observed. Variations in the positions of the furanose ring substituents correlate not only with differences in the geometry of the ring but also with differences in base-pairing geometries. These correlations would suggest a long-range effect of the furanose substitution on the interactions of the bases.

Analogues of nucleosides that lack the 3’-hydroxyl group are being studied extensively as potential therapeutic agents for the treatment of acquired immunodeficiency syndrome (AIDS). These compounds have been shown to be effective inhibitors of human immunodeficiency virus type-I (HIV-1), the causative agents of AIDS.14 Recently, 3’-azido-3’-deoxythymidine (AZT) was licensed by the FDA for the treatment of certain HIV infections. Other compounds that have shown promising activity in preliminary trials include 2‘,3/-dideoxycytidine and 2’,3’-dideoxya d e n ~ s i n e . ~3/-Deoxynucleosides are thought to function by blocking chain elongation, inhibiting the viral enzyme reverse transcriptase. The exact nature of this inhibitory mechanism remains to be determined. Differences in activity and molecular composition of the various analogues are not always consistent with a simple model of incorporation of the 3’-deoxynucleoside in D N A that would block the attachment of additional nucleotides. For example, the importance of the 3’-azido group in the activity of thymidine analogues is not easily explained. Full understanding of the requirements for effectively binding to and inhibition of the enzyme is required before a systematic approach can be taken to develop more effective drugs. This requires structureactivity studies, including correlation of geometric features of the various compounds obtained from conformational studies with activity and toxicity levels observed in vitro and in vivo. In addition, the role of other biological processes such as metabolism will need to be examined. In this paper we describe the solid-state conformations of A Z T and two of its analogues, 3/-azido-2’,3/-dideoxyuridine (CS-87) and 3’-azido-2’,3’-dideoxy-S-ethyluridine(CS-85). These compounds were chosen not only because of their structural similarities to A Z T but also for their promising biological activities. CS-87 is a potent anti-HIV-1 agent.6*7 M I C values for CS-87 and A Z T in ATH-8 cells are 0.40 and 0.32 pM, respectively. In peripheral blood mononuclear (PBM) cells, AZT is SO-fold more potent than ‘Medical Foundation of Buffalo, Inc. ‘University of Georgia. (Emory University and Veterans Administration Medical Center.

0002-7863/88/1S10-2277$0l.S0/0

,pt Y ,

N 8‘

AZT, R = CH3 (3-87, R = H CS-85, R = CZH, CS-87. However, the bone marrow toxicity levels of CS-87 are markedly lower than those of AZT.6 CS-87 is currently a strong candidate for clinical studies. CS-85 is also a selective anti-HIV-1 agent in PBM cells with reduced toxicity.* ( I ) DeClercq, E. J . Med. Chem. 1986, 29, 1561-1569. (2) Mitsuya, H.; Broder, S. Nature (London) 1987, 325, 773. (3) Mitsuya, H.; Matsukura, M.; Broder, S. In AIDS: Modern Concepts and Therapeutic Challenges; Broder, S . , Ed.; Marcel Dekker: New York,

1986; pp 303-333. (4) Herdewijn, P.; Balzarini, J.; Declercq, E.; Pauwels, R.; Baba, M.; Broder, S.; Vanderhaeghe, H. J . Med. Chem. 1987, 30, 1270-1275. (5) Broder, S.; Mitsuya, H. Proc. Natl. Acad. Sci. U.S.A. 1986, 83, 1911-1915. (6) Schinazi, R. F.; Chu, C. K.; Ahn, M. K.; Sommadossi, J. P.; McClure, H. Abbott-UCLA Symposium: Human Retroviruses, Cancer and AIDS: Approaches to Prevention and Therapy, Keystone, CO, April 1-6, 1987; J. Cell. Eiochem. 1987, 74, Suppl IID, paper P405. (7) Schinazi, R. F.; Chu, C. K.; Kahlon, J. B.; Shannon, W. M.; Canonico, P. G. 27th ICAAC, New York, Ny, Oct 4-7, 1987. (8) Schinazi, R. F.; Chu, C. K.; Feorino, P.; Sommadossi, J. P. 26th ICAAC, New Orleans, LA, Sept 28-0ct I , 1986.

0 1988 American Chemical Society

2278

J . Am. Chem. Soc., Vol. 110, No. 7, 1988

Van Roey et al.

Experimental Section (a) General Procedures. Procedures for the synthesis of AZT9 and CS-87’O have been previously reported. However, the samples of AZT, CS-85, and CS-87 used for this study were prepared by different procedures, which will be reported elsewhere. Crystals were obtained by slow evaporation methods. All three structure determinations were performed under identical conditions unless specified. X-ray diffraction intensities were measured on a Nicolet P3 diffractometer using Ni-filtered Cu K a radiation (A = 1.5418 A). The crystal was cooled with.a forced nitrogen stream. The temperature at the nozzle was 142 2 K. Measurement of the temperature at the site of the crystal prior to data collection indicated that the crystal temperature should be considered to be about 165 K. All data between 4 < 20 < 115O were measured. Six reference reflections were measured after every 120th measurement but showed no significant variations. The data with F > 3u(F) were considered observed, where u2(F) = (k/Lpcpl)[u2(r)+ (0.011)2]. Lorentz and polarization corrections were applied, but no absorption correction. The structures were determined by direct methods and refined by full-matrix least-squares, minimizing Z w ( F , - F,)2, where w = l/uZ(F) for the observed data and w = 0 for the unobserved data. Because of the limited number of data and the low thermal motion, only the exocyclic non-hydrogen atoms were refined with anisotropic thermal parameters. Hydrogen atoms were located in difference maps and refined in separate cycles after the refinement of the non-hydrogen atoms had converged. Two or three cycles of hydrogen atom refinement were alternated with two cycles of non-hydrogen atom refinement until the shift/error ratio in the first cycles of each set was less than 0.25. Atomic scattering factors and dispersion corrections were taken from ref.” Programs used include the data reduction package DREAM^^ and the direct methods programs MULTAN’.’ and SHELX84.“ (b) AZT (CIOHI3NSO4). The sample was recrystallized from a 1:l methanol-water mixture. The crystals are monoclinic, space group El, a = 17.538 (1) A,b = 12.021 (1) A,c = 5.6435 (3) A, B = 95.596 ( 8 ) O , V = 1184.15A3,Z=4,M,=267.25,D,l,= 1.499gcm-’,p=0.963 mm-I. A total of 1701 unique data were measured of which 1698 were considered observed. Final residuals were R = 0.037 and R, = 0.055 for the observed data and R,,,= 0.038; the standard deviation of an observation of unit weight was s = 5.995. All hydrogen atoms were located and refined. (c) CS-87 (C9HllNJ04).The sample was recrystallized from ethyl acetate. The crystals are monoclinic, space group P21, a = 9.7314 (8) A, b = 6.7517 (6) A,c = 8.3464 (7) A, fl = 91.854 (7)O, V = 548.09 A3, Z = 2, M,= 253.22, Dwld = 1.534 g cm-.’, p = 1.01 mm-I. A total of 827 unique data were measured of which 821 were considered observed. Final residuals were R = 0.046, R, = 0.065, and RaII = 0.046, s = 6.939. All hydrogen atoms were located and refined. (d) CS-85 (CIIH15NJ04). The sample was recrystallized from acetonitrile. The crystals are monoclinic, space group P2,, a = 5.567 (5) A, b = 22.093 (9) A, c = 21.149 (4) A, fl = 96.005 (9)O, V = 2587.24 AS, Z = 8, M , = 281.27, Dmld = 1.449 g cm-’, p = 0.908 mm-’. A total of 3627 unique data were measured of which 3567 were considered observed. Final residuals were R = 0.035,R, = 0.041, and R,,,= 0.036, s = 3.08 1. The azido group of molecule D is disordered. Two independent positions for N7’D and N8’D have been observed. Initially, occupancies and isotropic thermal parameters were refined until a suitable fitting of the observed electron density could be obtained. Occupanies of 0.55 and 0.45 were chosen and were held constant during refinement with anistropic thermal parameters. All hydrogens were located in difference maps. Atom H2’B of molecule A became nonpositive definite during refinement. Its B value was reset to 4.00 A2 and held fixed.

*

Results and Discussion Table I lists the atomic coordinates and isotropic thermal parameters (Eim)for the non-hydrogen atoms. Where appropriate, Bise represents the equivalent B calculated from the anisotropic thermal parameter^.'^ The numbering scheme, definitions of (9) Horwitz, J. P.; Chua, J.; Noel, M. J. Org. Chem. 1964, 29, 2076. (10) Lin, T. S.; Mancini, W. R. J. Med. Chem. 1983,26, 544. ( I 1) International Tables for X-Ray Crystallography; Kynoch: Birmingham, England, 1974; Vol. 4, pp 99-100. (12) Blessing, R. H. Crysr. Rea. 1987, I, 3-58. (13) Germain, G.; Main, P.; Woolfson, M. M. Acta Crystallogr., Sect. A : Cryst. Phys. Diffr. Theor. Gen. Crystallogr. 1971, A27, 368-376. ( 1 4) Sheldrick, G. M. SHELX84; University of Gottingen: Gottingen, F. R. G., 1984. (15) Willis, B. T. M.; Pryor, A. W. Thermal Vibrations in Crystallography; Cambridge University: Cambridge, England, 1975; pp 101-102.

torsion angles and conformational geometries were taken from SaengerI6 and are consistent with the rules of the IUPAC-IUB Commission on Biochemical Nomenclature.’’J* None of the bond lengths and angles are exceptional. The double bonds in the uridine rings are highly localized. The uridine bond lengths and angles do not appear influenced by substitution a t the 5-position. The azido group is not linear. The N6’-N7’-N8’ angle ranges from 171.0 (3) to 173.7 (3)’ in the AZT, CS-87, and first three CS-85 molecules. The azido group in the fourth CS-85 molecule is somewhat less well determined as the result of the disorder, but all values are within the range observed for other azido groups. The 28 observations contained in the Cambridge Crystallographic D a t a b a ~ e have ’ ~ N-N-N angles ranging from 168.0 to 176.7’ with an average value of 172 (2)’. The structure of AZTZ0has been determined previously, but coordinates were not available to us a t the time of this study and a structure determination under identical conditions as those used for CS-85 and CS-87 was thought to be beneficial for the comparison of structural details. The bond lengths, angles, torsion angles, and deviations from planes that are essential for the description of the molecular conformations of the seven independent observations (two for AZT, one for CS-87, and four for CS-85) are listed in Table 11. The molecules are shown in Figure 1. Dimers are shown to allow simultaneous illustration of the molecular conformations and the principal intermolecular interactions: the base pairing. Each pair of molecules represents two crystallographically independent molecules for A Z T and CS-85. The dimer for CS-87 shows two symmetry-related molecules. There is no a priori reason to assume that the difference in substitution a t the 5-position would significantly alter the molecular conformation. However, large variations in the conformational features are observed. Figure 2 shows all seven molecules superimposed by least-squares fitting of the uridine moiety. Assuming that differences in torsion angles smaller than about 15O are not important for conformational analysis, four different conformations are observed for the glycosyl link, two for the furanose ring, two for the C4%5’ link, and four for the azido group. Even the 5-ethyl substituent in CS-85 adopts two distinct positions. Only two of the seven molecules (AZTB, CS-85B) have identical conformations. AZTA and CS-85A differ from one another only in the position of the azide group. The glycosyl links (angle x) are all anti but span a range of 65O. Four furanose rings (angle v and out-of-plane deviations) are best described as having distorted twist conformations with 2’-endo. The others (AZTB, CS-85B, CS-8SD) have distorted 3’-exo envelope conformations. Average torsion angles and deviations of atoms C2’ and C3’ from the plane of Cl’, 04’, and C4’ indicate the large variation in the distortions. Four azido groups are in the trans position to the C2’-C3’ bond, one (CS-85C) is trans to the C4’-C3’ bond, and one (CS-85A) is 30’ out of the C2’-C3’ trans position. The last bond (CS-85D) is disordered. Two separate positions of nearly equal occupancy (0.55 and 0.45) have been determined, one cis with the C2’-C3’ bond and one 26O removed from this position. The orientation about the C4’-C5’ bond (angle y) is +sc (or gauche-gauche) in five molecules and s c (or gauche-trans) in the remaining two (AZTB and CS-85B). The uridine ring is not always perfectly planar. In molecules CS-85C, CS-85D, and CS-87 the average torsion angles of the (16) Saenger, W. Principles of Nucleic Acid Structure; Springer-Verlag: New York, 1984; pp 9-101. (17) IUPAC-IUB Commission on Biochemical Nomenclature (CBN) Eur. J. Biochem. 1970, 15, 203-208. (18) IUPAC-IUB Joint Commission on Biochemical Nomenclature, Eur. J. Biochem. 1983, 131, 9-15. (19) Allen, F. H.; Bellard, S.; Brice, M, D.; Cartwright, B. A,; Doubleday, A.; Higgs, H.; Hummelink, T.; Hummelink-Peters, B. G.; Kennard, 0.; Mothenvell, W. D. S.; Rodgers, J. R.; Watson, D. G. Acta Crystallogr., Sect. E: Struct. Crystallogr. Crysr. Chem. 1979, 835, 2331-2339. (20) (a) Gurskaia, G. V.; Tsapkina, E. N.; Skaptsova, N. V.; Kraevskii, A. A.; Lindeman,S . V.; Struchkov, I. T. Dokl. Akad. Nauk. SSSR 1986, 291, 859-862. (b) Birnbaum, G. I.; Giziewicz, J.; Gabe, E. J.; Lin, T.-S.; Prusoff, W. H., submitted for publication in Can. J. Chem. (21) Johnson, C. K. ORTEPII, Report ORNL-5138; Oak Ridge National Laboratory: Oak Ridge, TN, 1976.

J . Am. Chem. Soc.. Vol. 110, No. 7, 1988 2279

3'-Azido- 3'-deoxythymidine Analogue Structures

Table I. Atomic Coordinates (X104) and Isotropic Thermal Parameters (X102) for AZT, CS-87, and CS-85"

Bise

xla

Ylb

ZlC

1647 (2) 2599 (2) 2026 (2) 1328 (2) 2211 (2) 338 (2) 103 (2) -734 (2) -793 (2) -772 (2) 1123 (1) 2362 (1) -1178 (2) -1879 (2) -2521 (2) 1494 (1) 3268 (1) -159 (1) -138 (1)

9068 (4) 10258 (3) 10545 (4) 10095 (4) 11340 (4) 8960 (4) 8254 (4) 8527 (4) 9746 (3) 10596 (4) 9373 9514 (3) 7779 (4) 7888 (3) 7903 (4) 8444 (3) 10612 (3) 9908 (3) 10410 (3)

4033 (6) 6331 (6) 7867 (6) 7436 (5) 9871 (6) 5226 (6) 7286 (6) 7314 (6) 6483 (6) 8485 (6) 5577 ( 5 ) 4520 (5) 5594 (6) 5498 (4) 5283 ( 5 ) 2319 (4) 6512 (4) 5081 (4) 10202 (4)

184 (6) 175 ( 5 ) 195 (6) 186 (6) 243 (9)* 195 (6) 231 (6) 219 (6) 195 (6) 240 (8)' 186 ( 5 ) 195 ( 5 ) 290 (8)* 224 (8)* 283 (8). 237 (6)* 239 (6)* 220 (4) 287 (6).

3285 (2) 2425 (2) 3058 (2) 3733 (2) 2943 (2) 4596 (2) 5000 (2) 5817 (2) 5720 (2) 5557 (2) 3854 (1) 2596 (1) 6216 (2) 6914 (2) 7546 (2) 3383 (1) 1783 (1) 5085 (1) 5386 (1)

8271 (4) 6962 (4) 6706 (4) 7232 (4) 5870 (4) 8615 (4) 8340 (4) 8125 (4) 7686 (4) 6437 (4) 7995 (3) 7753 (3) 9224 (3) 9186 (3) 9257 (4) 8968 (3) 6573 (3) 8324 (3) 6087 (3)

-307 (6) -2654 (6) -4026 (6) -3476 (6) -5989 (6) -1 176 (6) 1296 (6) 779 (6) -1771 (6) -1914 (6) -1668 ( 5 ) -857 ( 5 ) 910 (6) 653 (5) 492 (6) 1320 (4) -2966 (4) -2879 (4) -4298 (5)

196 (6) 198 (6) 213 (6) 195 (6) 287 (9): 178 ( 5 ) 230 (6) 223 (6) 189 (6) 229 (8): 186 ( 5 ) 191 (5) 270 (8): 240 (8). 301 (9): 230 (6)* 276 (6): 194 (4) 299 (6)*

3188 (4) 4108 (4) 3464 (4) 2755 (4) 1956 (4) 3263 (4) 2369 (4) 1255 (4) 1919 ( 5 )

180 (6) 193 (7) 221 (7) 211 (7) 170 (6) 190 (7) 192 (7) 190 (6) 232 (8)'

11352 (3) 13306 (3) 6972 (3) 5869 (3) 4833 (4) 11525 (2) 15061 (3) 9611 (2) 9047 (2)

5282 3959 (6) 4189 (7) 4554 (8) 4711 (9) 1947 (6) 5849 (6) 6642 (6) 9593 (7)

2608 (4) 3873 (3) 1443 ( 5 ) 789 (5) 134 (7) 3088 (3) 4815 (4) 950 (3) 3435 (3)

178 (6) 179 (6) 308 (9)* 323 (9)* 553 (14)* 214 (6)* 261 (7): 196 ( 5 ) 250 (6).

12029 (3) 13928 (3) 13172 (3) 11931 (4) 9934 (3) 8865 (4) 7615 (3) 8249 (3) 8301 (3)

3633 5782 7431 7117 5021 4998 5833 7373 9475

(7) (7) (7) (7) (7) (7) (7) (7) (7)

Bi,

xla

Ylb

ZIC

6236 (2) 21 18 (2) 219 (6) 179 (6) -168 (7) 1741 (2) 2447 (2) 8736 (6) 191 (6) -2024 (7) 1456 (2) 6362 (2) 1608 (2) 233 (7) 10174 (6) 1373 (2) 1370 (2) 191 (6) -3370 (7) 1498 (2) 5898 (2) 223 (6) 11941 (6) 1855 (2) 324 (11): 1366 (2) 178 (6) -2273 (8) 7006 (2) 11984 (6) 2240 (2) 1996 (2) 852 (2) -4303 (8) 7102 (2) 326 (11)* 977 (2) 245 (9)* 13562 (7) 1892 (2) 1261 (2) 4823 (2) 1013 (2) -4456 (6) 277 (lo)* 190 (6) 15253 (7) 2431 (2) -7178 (6) 1129 (2) 4905 (2) 203 (6) 2972 (2) 196 (6) 10513 (6) 2636 (2) -7740 (7) 3442 (2) 496 (2) 254 (7) 4578 (2) 239 (7) 12753 (7) 2624 (2) -5494 (7) 165 (2) 4685 (2) 238 (7) 3667 (2) 277 (7) 3272 (2) 12920 (7) -5681 (8) -298 (2) 5213 (2) 311 (lo)* 3064 (2) 214 (6) 3630 (2) 12085 (7) -3191 ( 5 ) 1619 (1) 5321 (2) 209 ( 5 ) 2693 (2) 242 (9)* 3850 (2) 14069 (7) -90 (5) 2348 (1) 202 ( 5 ) 196 ( 5 ) 5652 (2) 2460 (1) 10429 ( 5 ) 2187 -7920 (7) 583 (2) 343 (lo)* 3914 (2) 184 ( 5 ) 1946 (1) 8705 ( 5 ) 1348 (2) 326 (10): 471 (12)* -9564 (7) 904 (2) 4125 (2) 3332 (2) 3754 (2) 11092 (8) 450 (12)* 279 (9): -10993 (8) 1188 (2) 4559 (1) 3555 (2) 11469 (6) 3696 (2) 255 (6)* 388 (10): -1267 (5) 2354 (1) 4673 (2) 11554 (7) 4986 (2) 4001 (2) 2352 (1) 229 (6): 1286 (5) 289 (7): 2859 (1) 6619 (2) 7326 (4) 1682 (2) 1026 (1) -3552 (4) 220 (4) 246 (6)' 658 (1) 4798 (2) 995 (2) 9912 ( 5 ) -6593 ( 5 ) 5734 (2) 316 (7)* 196 (4) 2682 (1) 10525 (4) 3222 (2) -7 (1) 7195 (6) 6791 (2) 352 (8): 3326 (2) 2552 (1) 215 (6) 3361 (2) 15549 (5) 2301 (1) 160 (6) 218 (6) 5598 (7) 5760 (2) 423 (2) 3380 (2) 3566 (6) 193 (6) 3313 (2) 228 (7) 7680 (7) 5607 (2) 3811 (2) 2178 (6) 748 (2) 3244 (2) 191 (6) 213 (6) 9334 (6) 6032 (2) 3967 (2) 389 (6) 273 (2) 2731 (2) 186 (6) 335 (11): 7916 (8) 4979 (2) 4086 (2) 306 (6) -86 (2) 6198 (9) 4862 (3) -1292 (7) 212 (2) 569 (16)' 3756 (2) 240 (9)* 4574 (3) 254 (9)* 10999 (7) 7085 (2) 3923 (2) -3033 (7) -315 (2) 231 (6) 3667 (2) 10082 (7) 7571 (2) -409 (2) 256 (7) 1679 (1) 168 (6) 4349 (2) 1683 (6) -831 (2) 260 (7) 1669 (2) 220 (6) 12148 (7) 7647 (2) 4882 (2) 3831 (6) 232 (7) 228 (6) 13362 (6) 7034 (2) 4908 (2) 2687 (7) -1425 (2) 1426 (2) 232 (6) -1405 (2) 269 (10): 1689 (2) 198 (6) 12395 (7) 6601 (2) 5373 (2) -1660 (2) 216 (5) 262 (10): 2354 (2) 9136 ( 5 ) 6618 (2) 3739 (1) 329 (7) 211 ( 5 ) -18 (2) 5570 (5) 1845 ( 5 ) 2264 (1) 173 (5) 6346 (2) 3148 (1) 13916 (7) 2820 (1) 4768 (2) 393 (11): 185 ( 5 ) 8126 (2) 3629 (5) 796 (2) -1398 (2) 287 (9)* 2441 (6) 14136 (24) 721 (1) 4254 (10) 358 (36)* 8247 (7) -1838 (2) 253 (8). 13371 (29) 458 (1) 4276 (8) 287 (39): 8438 (7) 1358 (6) 316 (9)* 14378 (17) 440 (7) -2215 (2) 3745 (4) 164 (2) 446 (24). 8429 (4) 8797 (7) 480 (2) 214 (6)' 12859 (24) 4981 (4) 3910 (7) 1896 (1) 702 (42)* 1101 (2) 3763 (1) 239 (6)* 6952 (5) 2501 (4) 3126 (1) 296 (7)* 7312 (2) 192 (4) 3946 ( 5 ) -772 (2) 3206 (1) 301 (7)' 5409 (2) 1669 (1) -410 (4) -1964 (6) 13019 (4) 2584 (1) 394 (8). -1578 (2) 4260 (1) 221 (4) 6811 (2) C(2C) -1506 (6) 5177 (2) 2126 (2) 192 (6) O(5'D) 9839 ( 5 ) 6586 (2) 5273 (1) 286 (7)* "The esd's are given in parentheses. An asterisk after the thermal parameter indicates that the atom was refined using anisotropic thermal parameters and the B value listed is B,.

2280

J. Am. Chem. SOC.,Vol. 1IO, No. 7, 1988

Van Roey et al.

Table 11. Conformational Parameters: Bond Lengths, Angles, Torsion Angles and Distances between Atoms from Planes That Represent

Important Conformational Features molecule AZTA N1-C1’ C 3’-N6‘ N6’-N7’

1.460 (4) 1.487 (5) 1.232 (4)

AZTB CS-87 CS-85A Bond Lengths (A) and Angles (deg) 1.502 (4) 1.477 (4) 1.467 (4) 1.494 (6) 1.480 (6) 1.483 (6) 1.248 (4) 1.214 (4) 1.220 (5)

N 7’-N 8’

1.122 (4)

1.124 (4)

C3’-N 6’-N 7’

C2-Nl-C1’-04’ (x) C2-N 1-C 1’